Understanding Statistical Significance versus Clinical Significance : From the Perspective of a Study, “Impact of Cell-phone on Human Health”

Abstract

Background & objective: In clinical research, statistical significance (p-value < 0.05) is often reflexively equated with clinical significance as well. However, statistical significance merely indicates the likelihood that a result occurred by chance, whereas clinical significance reflects the practical value of the finding to patient care. This study clarified the difference between the connotations of the two frequently used terms in medical research – statistical significance and clinical significance. Methods: The study utilized data from a PhD study (conducted during the period from July 2017– June 2022) involving 118 medical students. We measured hemodynamic variables (pulse, blood pressure, and parotid gland temperature) before and after a mobile phone call (talk-time intervention). Data were analyzed using paired-sample t-tests, Cohen’s d effect sizes, and correlation coefficients. Results: Post-intervention, local temperature increased significantly (97.4 ± 0.7°F to 98.0 ± 0.8°F, p < 0.001) with a large effect size (d = 0.80). Pulse rate also showed a statistically significant increase (89 ± 17 to 92 ± 17 bpm, p = 0.009), but with a negligible effect size (d = 0.18). No linear correlation was found between talk time and hemodynamic changes. Conclusion: The findings demonstrate that while pulse rate changes reached statistical significance due to sample size sensitivity, they lacked clinical significance. Conversely, local temperature changes demonstrated both statistical and clinical significance.

Ibrahim Cad Med J 2025, 15(1): 28-32